Ignition vs Glow - why the difference in power?

[Larry Renger]

My buddy Andy flies a model powered by a Forster 35 ignition engine.  He is using standard glow fuel.  When he runs the engine with a glow plug it puts out noticeably more power than when running ignition.

It would seem to me that the ignition system, which allows you to set the firing point wherever you want it, should actually increase power, not reduce it.

Anybody know what is going on?

[Willis Swindell]

I think that a glow would turn more RPM?
Willis 

[Alan Hahn]

My buddy Andy flies a model powered by a Forster 35 ignition engine.  He is using standard glow fuel.  When he runs the engine with a glow plug it puts out noticeably more power than when running ignition.

It would seem to me that the ignition system, which allows you to set the firing point wherever you want it, should actually increase power, not reduce it.

Anybody know what is going on?

My answer to your question is "no".

I would have thought that the spark would be better, but if it isn't then all I can figure is that the glow is giving a better burn than the spark. Why? maybe its in the details--weak spark, not advanced as much, glow is "on" for all the time,.......

[Andrew Tinsley]

Hallo there all,
                    A very interesting question! All things being equal (which they probably are not!) the calorific value of petrol is much higher than that of methanol, so one would expect it to be easier to cram in more petrol /air than methanol /air. Thus in theory you would expect the petrol to be the more powerful fuel. The petrol certainly makes the engine run hotter and in practice it is much easier to needle a sparkie on methanol.
  That is however, not the question being asked! Why would an old Forster run better on glow than on spark? . Normally I would not expect much difference. Being able to select the ignition point on sparks, would one expect, give a better performance. The glow is fixed ignition (depending on type of glow plug used), thus you might expect the ignition point to be optimum only if you selected the correct heat glowplug. Two explanations come to mind. The power in the spark is probably not enough to promote the best burn conditions. The small ignition coils used for our type of work are not very efficient. I have tried an ordinary auto coil for bench running (Used a transistorised system that limits the current through the points!) I have seen much higher revs on a given prop with this set up compared to the minature coils. The second reason is in the points themselves. The Forster points are very simple devices and will start to bounce above a given rev point. This will limit the revs very effectively! Points used on say the OK Super 60 and the Atwood Champion are basically automotive points and will take much higher revs than the type used on the Forster. In fact I use Datsun points on the OK Super 60!
  I have spent some time developing a microprocessor controlled ignition system. Among its many advantages is that it can recognise when points bounce and can ignore the spurious break when the points start to bounce. The micro controls a transistor which operates the coil low tension side.
  Other factors are the type of spark plug that you use, the vintage Champion plugs can degrade with age and give poor performance, replacing with a modern rimfire type might surprise you! A capacitor discharge system can also improve the engines performance, but gains are limited with the small model coils normally used. I have not been able to minaturise the CD system and it is bulky and weighs a lot. Try playing around with different coils, they do vary greatly in power output to the plug and this can have a significant effct on the performance of an engine. Hope this helps.

Regards,

Andrew Tinsley.

[dynasoar1948]

Hi Andrew,

Thank you for an excellent post.  I began in model airplanes early enough to run my first engine, an Ohlsson 60, on ignition.  The Arden glow plug was introduced a few months later and, being a beginning speed flyer, did not use spark ignition afterward.

There has been discussion of a vintage ignition speed catagory, which might get me to locate (or recreate) the points for my old Dooling 61, and build something.  I have been folllowing occasional columns in model magazines dealing with ignition engines and have seen circuits using a transistor to switch coil current.  Somewhere it was mentioned that the transistor circuit was rpm limited.  Are you aware of any modern circuits that would boost the performance of racing ignition systems beyond the 4-1/2 Volt circuits using automotive points and Smith coils, which were the standard in tether cars.  I recall that the tether cars were, for many years, substantially faster on spark than glow plug ignition.  I believe the proposed vintage classes would require breaker points, so optical or Hall effect sensors would not be legal.

A side light on Speed flying during the transition from spark ignition to glow plug.  The AMA rules of that time  dictated line diameter as a function of airplane weight (0.008-inch minimum with 0.001 added for each 3-oz over the basic 24-oz.), so elimination of the weight of an adequate ignition system reduced line drag of a typical 60 ship perhaps four MPH.

Regards,  William Stewart

[Charlie Pate]

My guess would be points.
Those of us that had the old cars remember the annual tune ups.
When the cars went to twelve volt ignition system things got better.
Point dwell (how long closed), resistance(dirty),all factored in.
Some hot rodders used dual (two sets of) points
The old NASH ignition used two coils.(car for you youngsters.)
T he main advantage of transistor ignition when introduced,was lack of mantainance required ,compared to points
 Also the rise time(the time it takes the voltage to build up enough to jump from one electrode to another)
with transistor ignition was far far better than with points.It was claimed that points rise time was slow enough
that it allowed the voltage to leak off a dirty plug while building up enough potential to jump the gap.
 Then there are spark plugs. I wonder how much research was done on Model spark plugs and their heat range
It makes a TON of difference in a two cycle. But thats another story 

[Andrew Tinsley]

Hi William,
             thanks for your praise, I probably don't deserve it! I have been working on my own and have had very little contact with the spark ignition community. My reported observations are therfore hardly the gospel!
  I must agree with you that the Smiths coil is far and away the best coil I have ever used amongst current and past manufacturers. It is heavier and more bulky than the other coils and I think that is the reason for its superior performance. It is more efficient and simply allows more power (in watts) in the high tension circuit, hence a more powerful spark. I have read somewhere that the Firecracker coil, was the coil of choice in days gone by. I believe that it was made by Smiths, perhaps you can throw some light on this?
  The transistorised ignition system that you refer to is now almost a de facto standard amongst many spark ignition flyers. It is a very basic circuit which simply reduces the current through the points to a few tens of milliamps, thus reducing wear and tear on the points. It will also work even if the points get smothered in oil. The basic switch is a TIP series transistor plus a few other components. I have used the circuit and it is totally reliable. I have never heard of the circuit having a speed limit. I have used it up to 12000 rpm with no obvious problems. It could be that when operating at very high revs, there may be a problem. The TIP series were (I believe?) designed as audio power transistors. If this is so, then it is possible that their gain will be getting less at high audio frequencies. This could account for the reported speed limiting properties of the circuit. I have used power FETs in place of the elderly TIP transistors, FETs have a much better high frequency performance than the TIP transistors, so I would expect no problems with this set up.
  If you are going to run your Dooling at very high revs (i.e. 20,000rpm region) you will probably encounter two problems. The first is points bounce, even with automotive type points. Apart from using my micro based anti bounce circuit (only ever reached the bread board stage!), I would suggest replacing the spring steel in the points with even stiffer spring steel!
  The second problem lies with the basic (Kettering) ignition system that we use. It relies on building up a magnetic flux within the coil and then suddenly collapsing the field by opening the points. It takes a finite time to build up the flux. As this time approachess the time taken to do one engine revolution, you can see that the flux does not have time to build up to maximum, before the points open. The higher the rpm, the worse this gets. So just when you need the best spark possible (for high rpm) the Kettering system simply starts to fall over. This is the reason that it is no longer used in high performance, full sized engines.
  The flux can rise more rapidly if the reactance of the coil is lower. It may be possible to design a coil with a lower reactance, for high speed use. My coil winder has now been cannibalised, so I can't wind such a coil, even if I could design a lower reactance coil! The othe factor which is overlooked is the capacitor which is wired across the points. This is said "to reduce the arcing at the points" This has beeen repeated so many times that it has reached the status of gospel truth, unfortuneately this is totally incorrect. The capacitor (usually around 0.1 microfarad) is actually an essential part of a resonant circuit. The value can be tuned to maximise the amplitude of the resonance and hence the maximum voltage that can be achieved. Its well worth experimenting here!
  I believe that Modelectric are now out of the coil business and their only US competitor has raised prices to astronomic levels. I also understand that some enterprising individual has cloned a modelectric coil and has had them manufactured in China (He beat me to it!!!!!). So it looks as if we no longer have a coil with the performance of say a Smiths coil. Maybe I should reclaim my coilwinder components from the latest project!

Hope that this has been of some use,

Regards,

Andrew Tinsley.
 

[Ralph Wenzel (d)]

Oops! DIHWIDT!  Post deleted.
Ralph

[Larry Renger]

Ralph, if you check the original post, Andy is using the same glow fuel for both ignition and glow.  The question is why glow produces more power on the SAME fuel.   I certainly understand the difference between alcohol and gas fuel in ignition engines, and that isn't the question.  Why does an accurately timed spark not produce as much power on a given fuel as a glow plug?

Now, Andy mentioned that there are car systems that provide multiple sparks for each ignition cycle to maximize power.  Basically, continuous fire as provided by the glow plug to activate the mix more thoroughly.  Would that work for our models?

Thanks everyone for great input, perhaps we can, collectively come up with an improvement in the performance of sparkers (not that I have one, but enquiring minds want to KNOW!)

[Steve Helmick]

Were the points removed when the glowplug readings were taken? I'd think there'd be a lot of mechanical drag... highly significant on a low performance engine...the smaller, the worse it would affect it. "Spark Scatter" or inconsistent spark timing while running, could easily reduce power 10% to 20%.

Regarding Andrew T.'s  and William S.'s posts...Dr. Spark's (aka Floyd Carter) transistorized ignition would be a huge step forward, especially if the points can be eliminated by using a modern trip system (magnetic, photocell, etc). I'm not sure if this is allowed by the current rules, however.  Steve

[dynasoar1948]

Hello Andrew,

 To complete my thoughts on spark versus glow ignition, tether cars, through the 'fifties, saw the magneto becoming dominant among the record holders.  Here, spark voltage increased with engine RPM.  I believe that magnetos were eventually obsoleted when resonant exhausts increased engine RPM to the present levels (over 40K in some examples) and glow plugs were sufficient.  I'm reminded that during the early racing  go-kart days, magneto points would float at high RPM, opening the way to what was termed "maveric spark'. This was an advanced spark resulting from E-gap/magnet interaction, which would build and advance with RPM.

If I build the Dooling vintage ship, a reasonable target RPM would be perhaps 18,500.

Will Stewart

[L0U CRANE]

Larry,

You may have seen my reminders, in this and some other forums, of two factors that probably have something to do with the question:

Gasolines have a much smaller range of fuel/air ratio tolerance, than methanol/ or ethanol/air fuels (by weight, for lab purposes, at least. )  Methanol-based fuels often include added nitromethane, which carries its own oxygen to a significant degree -not needing air ingested through the engine's intake processes- to produce added power. Methanol and ethanol are VERY tolerant of f/a proportions, and modern engines have vast intake and bypass channels to allow as nearly as possible a complete "fill' of the combustion cylinder. The alcohols can burn overly lean or rich, so they allow the engines to run way above torque peak - which is, almost by definition, the most efficient flow of fuel and air to the combustion chamber the engine is capable of...

Spark engines can only burn gasolines - within their inherent f/a ratio limits - in the amount of air the engine is capable of processing per rev. That tends to drop, past the torque peak RPM, faster, due to the narrow f/a ratio range, and so cannot add RPM to the horsepower equation as freely as methanol-based glow fuels. We can run methanol-base fuels at f/a ratios that would flood out gasoline (and let's not forget diesel: - kerosene-based fuels)

Add to that that most legal spark-ignited engines are from the late 40's, when power outputs like we see today were not even dreamed of...

Plus, there's that 50-year development span from then to now...

Our engines remain 2-cycle, air-pump devices, which are limited in what they can do by how much air they can pump through in each rev. Glows can "supercharge the fuel" with nitro. Schneurle porting can enhance volumetric efficiency over the "traditional" baffle-piston cross-flow pattern. These advantages help overcome the almost 2 to 1 difference in potential combustion heat yield between gasoline and either common alcohol...

Your original question was...?

[Andrew Tinsley]

Hi William,
             I have thought that a magneto would be a better bet than the kettering system for high rpm motors, the trouble is that they are quite heavy and I see no way to get the weight down. It has to be coupled mechanically to the engine and that is another problem altogether! It may be a good theoretical solution, but I have been deterred by the complexity, its fine for tethered cars, where the extra weight can be accomodated without too much trouble!
  Please be aware that the defacto TIP transistorised ignition circuit simply reduces the current through the points and avoids most problems with dirty points. It DOES NOTHING in the way of improving performance. I am not knocking the circuit, it does what it is intended to do, very well indeed and it is reliable. Its just that many people think it is some sort of magic and will improve performance, unfortunately that is wishful thinking.
  I am not sure about vintage speed, but in many old time events the use of points is mandatory. This is why I developed the  micro processor system, which can recognise points bounce when it occurs and ignores the spurious opening.
  To sum up Andy's problem. 1/ Use a new Rimfire plug 2/ swop your coil to a better one ie Smiths 3/ make sure that your points system is set up correctly and replace the spring  for a stiffer one. Try Woody Bartelt for a new spring (Aeroelectric) I would not bother with the transistorised ignition, it will not make the engine turn any faster, despite its other advantages.
  Another post referred to using dual points, this really is a non starter, you would need two sets of points and coils, not really an advantage for sports or competition flying. The contributer is quite correct in that multiple firing of the plug during combustion can improve the performance of an engine. I have borrowed a system that does just this and tried it on various old spark motors, it DOES improve performance a little on most of the engines I tried it on. However, this system was developed for racing two stroke motorcycles and there is no way that it could be minaturised for model aircraft use.

Regards,

Andrew Tinsley

[EddyR]

Point chatter,float, bounce,whatever you want to call it was a big problem back in the 60's on drag cars. I ran three stock and pure-stock drag cars back then with duel points and chatter became a problem at around 5000 RPM. In the stock class we could not run solid state ignition.I consistently ran my 68 383 Road Runner at 103 MPH in the 1/4 mile.I held national record for three months that year. One time I sneaked buy with my capacitor discharge system and ran 105 MPH as I was able to shift 500 RPM higher. I ran those big blocks at 5200 RPM. I also had a 68 street Hemi and the best shift point was a low 4600 RPM.Those were the good old day.
  I ran one of the Forster's back in the 50's and all I remember is how bad it beat my fingers
EddyR

[Larry Renger]

Again, Andy is running the same Glow fuel in both spark and glow modes.  The timing unit is installed at all times.  The ONLY difference is the use of a glow plug vs spark plug.

[Andrew Borgogna]

Hi guys I am the "Andy" Larry has been referring to.  I am currently using a Don Huddelson (I hope the last name is right) transistor ignition.  Strange thing about this circuit it does not require a capacitor (condenser for you old people) to function.  I had a chance to talk to Don about that at VSC and he said having one made no difference at all.  My understanding of the Kettering system was the cap is what provided the fast discharge path for the primary when the points opened.  Slow charge fast discharge.  I admit I am no expert on ignition but the issue with the primary of the coil not having enough time to fully charge before the points open coupled with the inefficiencies of the coil we use seem to me to be the most likely candidates for the problem.  Larry is correct the Forster .35 with all things equal except the mode of ignition runs much better on glow.

One thing that nobody has mentioned (or I just missed it) is the difference it the amount of advance between the two modes.  I seem to remember that glow runs around 40 degrees before TDC.  Yes I know that everything from nitro content to humidity affects this number but it is somewhere near 40 degrees.  I have no idea what the Forster is timed at at full advance, it would be simple enough to determine.  It maybe that the O/S A1 plug (hot) + 10% Nitro might produce more advance leading to more power.  Just a thought.  Like Larry I too have enjoyed this thread, thanks to all who have participated.
Andy Borgogna   

[Alan Hahn]

Hey Andy,
I mentioned it!! (the timing not being as advanced in the sparker---at least I think I did! )

But since I am sure your Engine isn't turning at 40krpm, the "only" explanation (in my mind anyway) is that the glow is burning the fuel---in some way---more efficiently or with more "bang" than the sparker.

[Jim Kraft]

Andrew; The condenser does no good in a transistor driven circuit because all the points are doing is triggering the transistor. The transistor does the job of the points in breaking and closing the circuit to the coil. I don't know why your engine runs better on a glow plug, as all of my ign. engines run better or as good on spark as they do on a glow plug. I have a couple that run way better on spark. The engines I have that run almost the same on either are high compression engines, and maybe that is the difference on those. I do have a Forster 29 rear valve which uses the same timer as your 35. The timer does have a stop so you can only advance it so far and I run mine all the way full advanced with the timer arm up as high as it will go. If your engine will not run well with the timer raised all the way it may be that it is not fully broken in. You might also want to pull the prop and prop washer and look to see if the timer is moving all the way on its slot. The point gap should be around .015" and this also effects timing to some degree. I run mostly Modelectric coils, but also run some Aerospark coils. Both seem to work fine. As mentioned before, the Rimfire plugs are way better than anything else I have found. It is interesting to note that my Anderson Spitfires have a stop that is set at .004" between the point arm and the stop when the points are open to keep them from bouncing even though they don't want the engine run over ten thousand RPM. I have never had a problem however with the Forster points bouncing and I have run it pretty fast on small props. I am still learning about spark ign. motors even though I have been at it for about 10 years now. But boy they sure are fun.

[Andrew Tinsley]

Hi everyone,
                  Just a very quick post. We are, or should be explaining why a Forster 35 runs better on glow than sparks. Lots of serious expositions, not the least from me!
 I have just dug out one of my Forster 35s, running it on 5% nitro and around 25% castor fuel, I have checked it out on both spark and glow. Guess what, I get approximately 10% better performance on spark than on glow! I suggest that Andy looks at a decent plug like a Rimfire and a Smith's coil, even on a Modelectric coil, I am seeing a 7% difference.
  That should muddy the water!

Regards,

Andrew Tinsley

ps My comments on the capacitor being a vital and tuneable component, went for the non transistorised ignition systems. If you don't believe this, try running without it and see how far you get!

[Larry Renger]

Thanks for the test, Andrew T., a most interesting result!   I think Andy B. is running 10% nitro, and thus has more "advanced" timing with the engine running on glow, but the spark setting is the same no matter what fuel is used.   Could you re-run your engine with hotter fuel and see if the result is the same?

[raglafart]

A very interesting thread, again I would also like to thank all the contributors, especially Andrew Tinsley.
I've only just dipped my toe in running spark ignition motors and have to say what a joy and pleasure it's been so far, I hasten to add that it's still early days for me though.
I'm am currently using a Modelectric coil, TIP 42 circuit, NGK ME8 plug and 3 Nimh 900 mah batteries to run my Orwick 64. I'm using avaition grade gasoline and Castrol M in a 3:1 ratio and also tongue muffler which works very well.
I'm really impressed with the power and running characteristics of the Orwick, must put a tacho on it next time out, but it swings a 13-6 Zinger cut down to 12-6 with authourity
From what Andrew was saying it sounds as if the power loss could very well be in the coil, plug or points and I'd ask Andy what he's using for those and also what type of batteries he's using?
Very interesting topic

Cheers John

[Andrew Tinsley]

Larry,
        I don't run any 10% nitro, so don't have any to hand. I will buy a pint next time I get to town, but it won't be for a couple of weeks. Will report back with results when I get them. I don't expect to see agreat difference percentage wise, although I would expect higher rpm.
  Can you tell me what Andy is using as far as plug and coil ? I don't know if it has been mentioned already, if so please forgive me.
  Just as an aside, I was a bit surprised that I got a 10% difference, its a bit higher than I would expect from past experience. Thats what makes sparkies so interesting, they are just plain awkward sometimes!

Regards,

Andrew Tinsley.

[dynasoar1948]

Andrew,

Do you recall the Clevite "Spark Pump" of the early sixties?  While this ultra simple piezo-electric device appeared on a few Clinton utility engines. now it seems that the only application of this phenemon is in cigarette lighters (and "earthquake light" as I can testify to from the '94 California event).

Regards,   Will

[Andrew Tinsley]

Hi All,
       Just neeed to correct my last post, the engine used was a Forster 29 not 35! I didn't even know that Forster did an ignition 35. I have a later 305, that is a spark engine, was this the type that Andy used for his title post?
  Secondly, Will, I have never heard of the Clevite "spark pump" of sixties vintage. Its something that I will not readily admit too, but I have tried to build a piezo electric ignition system using either salvaged cigarette lighter elements or those used to ignite gas barbecues. To put the very best face on it......I did manage to actually get an engine to fire once! Its interesting that you mention it, I suppose it must have worked back then. The main problem is that you can get the volts, but the current is so low that the actual power in the spark is very low indeed (according to my back of the envelope guesstimations..er I mean calculations). Not good for our purposes! I suppose that if you cut the crystal differently then it could work, but it got beyond me! I did try to boost the current, by using a charged capacitor, but after destroying a 25 kilovolt rated capacitor, I gave up. I know when I am beat!!!

Regards,

Andrew.

[Jim Kraft]

Andrew; Talking about current made me think of something else. I run no smaller than 18 ga. wire for the battery, transistor, coil circuit. I like to run 16 ga. and with a short run using the transistor, there isn't much loss. I think this is one of the reasons I get so much better spark with the transistor than I do with a capacitor and standard ign. I also run three AA 2500ma. NIMH batteries that will get me through at least twelve flights without recharge. My ign. circuit is very simple with one large 10 amp audio transistor and a 22 ohm resistor to the points. I probably get some loss through the transistor, but not through the short length of wire. I am sure that Don Hutchinson's, and Floyd Carter's ign. system is better, but this is working OK for me.

[Charlie Pate]

Hi All,
       Just need to correct my last post, the engine used was a Forster 29 not 35! I didn't even know that Forster did an ignition 35. I have a later 305, that is a spark engine, was this the type that Andy used for his title post?
  Secondly, Will, I have never heard of the Clevite "spark pump" of sixties vintage. Its something that I will not readily admit too, but I have tried to build a piezo electric ignition system using either salvaged cigarette lighter elements or those used to ignite gas barbecues. To put the very best face on it......I did manage to actually get an engine to fire once! Its interesting that you mention it, I suppose it must have worked back then. The main problem is that you can get the volts, but the current is so low that the actual power in the spark is very low indeed (according to my back of the envelope guesstimations..er I mean calculations). Not good for our purposes! I suppose that if you cut the crystal differently then it could work, but it got beyond me! I did try to boost the current, by using a charged capacitor, but after destroying a 25 kilovolt rated capacitor, I gave up. I know when I am beat!!!

Regards,

Andrew.

Having seen the Clinton service sheet on this type ignition , I guess there was one!
Inever did see one though.
As I under stood it, the crystal was thumped by a spring powered lever or plunger thus producing the energy.
I just couldn,t believe such a thing existed:much less was manufactured.
I THINK thihs is the same thing used on the propane grill ignitor.
Stick you hand near the gap and hit the button if you are into pain and self abuse(definatly NOT reccommended..

[Andrew Tinsley]

Hi Jim,
         I was a bit surprised to read that you got better performance with the "standard transistorised system" Than you did with the old points and condenser set up. My first thoughts was you had not tuned in the condenser value to what is a resonant circuit. So I set up my tuned capacitor / coil on an engine and then swopped for the transistorised system. You are RIGHT! the transistorised system did give more revs. So egg on my face I am WRONG. This shows that when you work essentially alone you are likely to make incorrect assumptions. That is what is good about forums like this, people can point out the error of your ways! There is no doubt that tuning the condenser on the old fashioned circuit does work. I have used a storage ossciloscope with high voltage probe to measure the voltage on the condenser. You can tune it to a maximum and this can be anything up to 1000 volts. This is why you get such a high voltage on the secondary with a turns ratio of only around 50 or 100 to 1 depending on the coil. The 3.5 volts on the primary would only give you an eht of 350 volts maximum on the secondary if this wasn't true.
  There is no condenser on the transistorised unit so how come this set up is so good? Might be worth thinking about! I have some ideas as to why it works, now that you have made me think! But I am cautious about going into print and making another incorrect statement. I shall do some investigating and get some hard facts before saying any more!
 
Here is one apologetic,

Andrew.

[Jim Kraft]

Andrew; I have made so many wrong assumptions with this stuff myself and then had to rethink the whole thing. You are right on about these forums, as there is a lot of info that can be traded, and the learning continues.

So wipe the egg off your face as I have seen two people run the same tests and come up with different answers.  I for one really enjoy your posts, and have learned a few things from you also. When things are looked at from different perspectives, it makes us all think, and learn new things.

[L0U CRANE]

>>Hi guys I am the "Andy" Larry has been referring to.  I am currently using a Don Huddelson (I hope the last name is right) transistor ignition.  Strange thing about this circuit it does not require a capacitor (condenser for you old people) to function.  I had a chance to talk to Don about that at VSC and he said having one made no difference at all. 

Andy,

That's Don Hutchinson, and he's the guy who converted some modern (mostly OS?) engines to spark for OTS years back. Way too reliable and easy to use, as well as dependably powerful. And that's why (most) OTS rules restrict the points bonus to authentic older engines, or replicas that hold true to the conditions of the original engines...

The capacitor, aside from the possible rapid discharge factor*, served to absorb voltage spikes across the breaker points. Points and gap settings can last a bit longer if you are not burning pits in one surface or the other.

* - The 'rapid field collapse' of the coil field when points open is not affected by a capacitor buffering the arc at the breaker points. The transistorized switching charges the coil and allows its field to collapse rapidly by arcing at the plug gap...

There are a couple of transistorized spark circuit boards available. Joe Klaus used to sell what he called a Schmidt trigger - one transistor mounted on a small strip of balsa for gluing into the model, good instructions for installation, no condenser. "Dr Spark" occasionally visible in several CL sites, used to sell  a small circuit board - discreet components assembly - potted in a silicon or other protectant - that worked excellently, too. As emphasis onhow little current passed through the breaker points, the wires to that area for +, - and ground were very small. Signal mattered, not raw current as in condenser supported spark systems.

Both CL OTS and SAM mention breaker-point spark as eligible, not Hall Effect systems. I Think...

[pipemakermike]

I have one of Dons contactor units fitted to my LA46 in a Go Devil.  This combined with a Dr Spark transistor unit has powered the model to 3 wins so far.

[raglafart]

On cleaning down my Orwick after its pile driver imitation (Sadly my Super Duper Zilch is no more!) thankfully no major damage. I was surprised to see the ratio of points closed to point open during a revolution. Why are the points open longer than they are closed, does it take that long to collapse the field and generate the spark? I would have thought that the points needed to be open less time than closed to charge the primary winding, what's going on here?
Cheers John

[Tom Perry]

John,

Too much of anything ain't good, but it ain't always bad.   

I don't presume to be an expert on Sparkies or ignition circuits.  I do have basic knowledge of electronics and will try to answer your question.

Since I'm a simple guy I'll try to give you a few simple answers.  I feel sure that Andrew Tinsley, among others, will point out any errors, or possibly add to, or explain it better than I.

The Kettering ignition system is one of the most brilliant systems ever devised, my opinion.  Like many other useful devices compromises were made to allow it to operate in the most efficient manner.

It takes a certain amount of time for a coil or inductor to charge up, once you exceed that time you are just wasting energy from the battery.  Not a good thing for a small battery with no provision for recharge.

It also creates a certain amount of heat from the current flowing through the rather low resistance of the primary windings of the coil.  Not a good thing as it may melt and open or short through the insulation to ground or to the secondary.  Either way ruining your ignition system and probably your day.   

In an automobile with multiple cylinders too much dwell time is not usually a problem.       Once you get beyond a V-eight oft times multiple points and coils are used.  Even a high performance (revs)  4-banger would require the multi points/coils.

The two stroke model engines require a spark once per revolution the four stroke only needs only half that so the distributor on a typical auto engine needs only half the revolutions of the two stroke engine.

I've never heard of a four stroke ignition engine for models, although there probably is one.  If I built one though I would probably not use a distributer as it wouldn't hurt anything to let it spark on the exhaust stroke, other than wear out the plug faster.  I said every thing is a compromise didn't I?   

[FLOYD CARTER]

Those of us who run spark ignition engines in C/L stunt aren’t too concerned with squeezing the last bit of RPMs from our engines, unlike the free flight people, who always want more!

What we may not realize is that the maximum RPM for any given engine is established by the manufacturer by machining the points cam!  Why is this so?

First, realize that for a given dwell angle, the dwell time decreases with an increase in RPM.  Dwell time can ultimately become too short at high RPM for the coil to completely saturate.  When the points first close, full battery voltage is impressed across the coil primary winding.  But the coil current does not flow instantly.  The current build-up, and hence the magnetic flux increases at an exponential rate (but a straight line increase can be imagined without too much error in the calculations).  Unless the dwell time at high RPM is sufficient to fully saturate the coil’s magnetic circuit, the points will open too soon, and the coil energy will discharge before it has reached full “charge”.  This will result in a weak spark and “missing” or sputtering.

To put some numbers on the problem, I’ve measured the dwell angle for several spark engines:

O&R 60                 80 degrees
Orwick 64              90
Arden 19               120
Brown D                 70
O&R 23                 115
Cunningham 65       95
Arden 09                150

Also, I have measured the time it takes for a Modelectric coil and 3.6 volt NiCd source to fully saturate.  This is 0.0018 sec.

Using the equations

          0.0018/T(sec) = DA/360
and
         60/T(sec) = RPM

where T(sec) is the minimum dwell time required
           DA = dwell angle, measured

solving for RPM gives:

O&R60                  7407 RPM
Orwick 64              8328
Arden 19              11110
Brown D                6486
O&R 23               10657
Cunningham 65     8796
Arden 09             13888

So, running these motors past their most efficient spark RPM will cause degradation of spark, even though the engine, for the most part, will tolerate the stress of higher RPM.  Spark performance can be increased by modifying the points cam for longer dwell angle (filing or grinding is required).  Alternately, a higher battery voltage will promote faster build-up of coil magnetic flux.  One must be careful using anything over 3.6 volts because coil current gets very high in a big hurry!  Even at 3.6 volts, the peak battery current is 4.5 Amperes!  Leaving the battery connected with points closed for more than a few seconds will overheat the coil.

One last observation.  I have run spark engines using gas/oil and glo fuel (always with spark ignition).  I have found no difference is performance between these fuels.  Gasoline has a higher specific heat than methanol, but the fuel/air ratio is less for gasoline, meaning that the two fuels seem to compensate and balance out about equally.

Floyd Carter (Aero Ply Research Co.)

[Tom Perry]

Floyd,

I didn't quite understand your formulas:

0.0018/T(sec) = DA/360
and
60/T(sec) = RPM

as presented but I think I understand how to solve for the most efficient RPM for a given dwell and saturation time for the coil.

60/time in seconds will always equal 33,333.3333 RPM for the coil you measured.

So if we take the dwell angle as measured and we divide it by 360 degrees we compute a percentage of the total angle of 360 degrees.

in the case of the Arden 19    we divide the dwell angle of 120 by 360 and find that it equals 33 and 1/3 percent or expressed as a decimal   .3333333.

so we take our previous computation of 33,333.33 RPM multiply by .3333 and the result is 11,111.11 RPM.

so if we let ST = saturation time and DA = dwell angle as measured  the following formula would compute,  MERPM or maximum efficient rpm.   

60/ST * DA/360 = MERPM

(60/.0018) * (120/360) = 11,111.11

33333.33   *    .3333   =  11.111.11


DAMMIT Floyd,  You and Andrew Tinsley have really pissed me off!  You both, no pun intended, have sparked my interest in your little sparkling toys.   

[FLOYD CARTER]

I used  the ratio proportion equation because it is the simplest to understand.

The ratio of measured minimum dwell time (0.0018) to time of one revolution equals the ratio of measured dwell angle to 360 degrees.

and RPS is the recriprocal of time for one revolution.

Finally 60 X RPS is RPM

I omitted the step of solving the three simultaneous equations for RPM.

Your numbers are slightly diferent from mine.  I rounded off for simplicity.

Floyd in OR

[Tom Perry]

Floyd,

How did you measure the saturation time on the coil, Oscilloscope?

What means do you use to ensure the engine does not stop with the points closed?

Can the engine be stopped by using a second switch in series?  Would it be competition legal to do so? 

[Bryan Higgins]

Thank you for all that information.  I learned a great deal i did not know.
What a great site filled with wonderfull people.
                                                                           Bryan

[FLOYD CARTER]

Tom Perry:

Yes, I set up an oscillator (a 555 timer chip) so that I could vary the ON time, and the interval between pulses was about 20 msec (but the frame rate doesn't matter).  With the oscilloscope I could watch the current build-up each pulse (that requires a small-value current sampling resistor in the circuit.  The waveform is a rising exponential wave, and I measured from the 10% to the 90% current value (the usual definition of "rise-time")

With a propeller mounted, and flying in genuine air, the prop always stops at beginning of compression and stays there.  Points are closed at that position.  Free flight guys must terminate engine run with a timer-switch, and so no further current flows.  On C/L stunt, we have full battery current flowing with engine stopped and during the glide to a landing.  I'm always trying to land close to my helper, who is instructed to head for the main power switch ASAP.  The 10 or 15 seconds glide/landing doesn't seem to cause any big trouble.

A switch must be used in the battery lead, and it must be capable of switching at least 5 Amperes.  A flight run timer can be wired in the points circuit if using a transistor coil driver, because the operating current there is just a few milliamps, and the smallest switch will do.

Hope this answers your questions.

Cheers,

Floyd Carter

[raglafart]

Maths was never my strong field, I struggled with quadratic equations and boolean algebra, it was like a foreign language to me. Thank you for the explanation Floyd. I have at least a basic understanding of what is going on now and won't be messing with the points cam anytime soon!
I'm taking 3 omega 3 capsules with every meal these days, but I think it's a wee bit late to look for much in the way of brain power improvements at my time of life. Just wish I'd eaten more fish as a youngster
Boy I'm kicking myself for putting in the Zilch, ah well, they come and they go.
Cheers John

[Jim Thomerson]

Talking about go-fast.  I recall an article in Flying Models (I think) back when spark ignition was still being flown in CL speed.  Title was something like We Flew 190MPH.  My magazine was incomplete. But there was a picture.  it was a Dooling 61 with a magneto running on pure oxygen.  So I guess a spark engine can be made to run pretty fast.

[FLOYD CARTER]

Jim.  I think 190 MPH with spark ignition is wishful thinking!  I remembe3r Don Newberger flying a speed model with a McCoy 60 on spark, and it might do 120 MPH on a real good day.  But that was about 1948 or so, and breaking 100 MPH was a big deal.

Floyd in OR

[Don Hutchinson AMA5402]

Re: The 190 mph run, it was done with a tank bleeding pure oxygen into the engine. Immediatly resulting in a rule making gaseous boosts illegal. I worked with a guy at TRW who claimed he was the one that did it. He was a good spacecraft mech and quite capable of doing so but I had never seen his name in any model mags so who knows???
Don

[dynasoar1948]

Floyd and Tom,

A few comments on the points you've introduced (no pun intended).  Ray Arden was first of all an ignition specialist, and wound coils for Glen Curtiss thirty years before introducing his first model engine (Mighty Atom .098).  He also made what were then (late thirties) the smallest engines in history: I believe that one was a 0.008^3".  While these were ignition engines, he developed the model the glow plug to simplify his mini-engines and introduced it to the industry in 1947.  My first control line speed engine was an ignition Arden .199, which I completes and flew just before the glow plug arrived.   Incidentally, I flew the ignition ship on glow, with the ignition system still in place but inop.  A couple of MPH faster, which I attributed to using Arden Glo-Flite #1, with nitromethane and nitrobenzene, in place of the methanol, castor oil, amyl acetate fuel I had run previously.

I'm curious as to why dwell is such a small portion of the total angle.  Why not extend to 270 or 300 degrees?  I dont have points for my Dooling 61, but believe that dwell was at least 180 deg.

I recall the 192 MPH claim in Flying Models Magazine in late 1949.  I believe the author was a well known modeler named Lou Garami.  The Dooling would have had to develop about 1.75 times the HP of the 159 MPH George Fong "HellRazor" flight made the previous September at Trenton N.J. (I was there, flying A Senior).  I question whether the Dooling would have survived at that power level long enough for an official length timing and the size of the pressure tank even at the claimed 800 PSI, looked way small for the volume required.  Assuming  it happened, the engine would have had to push BMEP up at least 60%, pulling a higher pitched prop, since there just was not much additional RPM to be had.  Oxygen boost would raise BMEP so maybe....   Nitrous oxide would have been easier.though.

Will Stewart

[dynasoar1948]

I believe that the fastest official ignition speed flight was about 143 MPH, in early 1948.  Airplane was powered by a Dooling 61.  Ignition tether cars eventually went faster than this, both with Kettering and magneto sparks.  A couple of weeks later a glow plug Hornet did a 153 official.

Will Stewart

[Andrew Tinsley]

Hi there all,
                I have not been following this thread over the last week or so. We have had good weather in the UK and my wife had plenty of jobs for me to do outside!
  Floyd has made some extremely good points re the dwell angle. I very briefly mentioned this in my first post. The weak point of our miniature Kettering ignition system is that it takes a finite time for the magnetic field in the coil to build up. For high revs, the dwell time of the points is insufficient to build up the maximum magnetic field so as the revs go up, there comes a point where the spark volts and energy begin to decrease, just when you need them! The revs are limited by the ignition  system and hence I am very dubious of some of the very high speeds that were claimed. It doesn't matter how potent the fuel brew, if the spark is almost non existent, then the engine wont rev! Suppose that one could use very high pitches on the prop (higher than normal for speed!) and even if rev limited by the electrics, then you just may get near some of the huge speeds quoted! There is another way to delay the onset of this problem and that is to build a coil with a lower reluctance value. This will shorten the time to get to maximum core flux and hence delay the nasty area where the spark begins to fail  Floyd's review of dwell angles on various engines was very useful for me, I have been looking for this type of data for a long time! I have only been able to compile it using the engines I happen to own.
  I am very reluctant to disagree with Floyd on one point, after all he is a guru when it comes to spark ignition! Floyd talks about the coil becoming saturated. If by that he means that the maximum field possible for a given primary current, then I have no problem. However "saturated" has a precise meaning and that is the maximum field possible from the core no matter how big the applied current. Most people may remember the school experiment plotting B versus H and getting the characteristic s shaped hysteresis curves. I have done this for cores that I have used to produce my  home made coils. The typical currents quoted by Floyd (and measured myself) only give around 30% of the saturated field. I have heard and read this quotation that the coil saturates. It is repeated so often (even seen it in text books!) that it has acquired the status of "truth". I am sorry, but it simply isn't so. I have even done the experiments with a dud Modelectric coil and got the same answer. A simple experiment will show that the coil is not saturated. Measure the maximum spark gap width for a system running on say 3 Ni cad cells. Then add a fourth cell and repeat the measurement. The maximum gap that the spark will jump is now bigger. Be quick and don't overheat the coil! Just consider, if the core was saturated on 3.6 volts. then increasing the voltage to 4.8 volts should make no difference to the spark gap, IF the coil was saturated. In fact you will always find that the maximum gap will increase with increased voltage. A sure sign that the core / coil is not saturated.
  I need a little help, the way the voltage and current in the coil behaves during a complete cycle is very much more complicated than many people assume. I have a large number of sketches of storage sope display for various set ups and I would like to copy some of those sketches to this topic. Can anyone tell me how I can do this??
I find that my contributions are very long winded (sorry for that!) and a few sketches replace a large number of words! I would like to show why the correct value of capacitor is very important. We are lucky in that most people use a 0.1 microfarad capacitor, which is about right, optimising it gives control of getting the highest theoretical secondary voltage and a longer duration spark. I am still trying to figure out why the transistorised system gives  a better output!

Regards,

Andrew.
 

[FLOYD CARTER]

Andrew.  You are correct.  I have tossed an incorrect term in my analysis.  Of course, the iron core is NOT saturated, or even close to saturation.

I should have explained that the maximum possible coil primary current (and hence the accompanying magnetic flux) is simply governed by Ohm's Law.

The Modelectric 3-volt coil (and most others) has a primary winding resistance of 0.8 Ohms.  So with a battery of 3.6 volts, the limiting current is just 3.6/0.8 = 4.5 Amps.  It can't go any higher.  That is what I meant by saturation.

Floyd in OR

[Andrew Tinsley]

Floyd,
         Thank goodness you made a slip of the tongue re saturation! I must admit that I was a bit bothered about taking you to task. I am only a newcomer to  vintage spark ignition engines and I know you have been at it for perhaps more years than you care to remember!
  As a matter of interest, I have an incurable muscular degenerative disease and I was told a couple or so years back, that I would be in a wheelchair soon. That was a blow....... no more control line stunt flying! So I decided to take up vintage sparkies and old time aircraft under radio assist, so I could keep flying. I think the challenges of spark ignition have done me a power of good, it certainly seems to have slowed down my illness. However I have not got in enough stunt flying!! Too much fiddling with spark ignition!
  Tom...... I am really sorry if I have pissed you off!!!!!!!! Spark ignition does exactly the same for me too! Although I am a physicist with a good electrical background, there are some aspects and phenomena that I really am at a loss to explain. It can get very frustrating at times. I will try and get some sketches, into my next contribution, showing the sort of behaviour that I saw on my test bench. Some I can understand, but others leave me struggling for an explanation. Hope someone can see the light.

Regards to all,

Andrew.
 

[raglafart]

My friend just returned my book by Bob Shores "Ignition coils and magnetos", for anyone who has been reading these pages will find this book very interesting if they are like me just learning about spark ignition.
Cheers John